2006-1830: FIELDBUS NETWORKS TOPIC IN INSTRUMENTATION AND CONTROL SYSTEMS COURSES

Sri Kolla, Bowling Green State University Sri Kolla is a Professor in the Electronics and Computer Technology Program at the Bowling Green State University, Ohio, since 1993. He worked as a Guest Researcher at the Intelligent Systems Division, National Institute of Standards and Technology, Gaithersburg, MD, 2000-‘01. He was an Assistant Professor at the Pennsylvania State University, 1990-‘93. He got a Ph.D. in Engineering from the University of Toledo, Ohio, 1989. His teaching and research interests are in electrical engineering/technology area with specialization in artificial intelligence, control systems, computer networking and power systems. He is a senior member of IEEE and ISA.

Joseph Mainoo, Bowling Green State University Joseph Mainoo is a graduate student in the Master of Industrial Technology degree at the Bowling Green State University, Ohio. He received his B.S. in Electronics and Computer Technology from the Bowling Green State University, Ohio, in 2004. He also has a Diploma in Management Information Systems from the Institute for the Management of Information Systems, London, UK. His academic interests are in the areas of information technology and electronics. He is a student member of ISA. Page 11.642.1

© American Society for Engineering Education, 2006 Fieldbus Networks Topic in Instrumentation and Control Systems Courses

Abstract

Fieldbus networks are digital, two-way, multi-drop communication links that are used to connect intelligent control devices. These are currently introduced in the industry to replace the traditional 4-20 mA point-to-point connections. It is important to integrate fieldbus networks topic in technology courses to align the curriculum with the current industrial practices. This paper, therefore, presents how the fieldbus networks topic is integrated into ECT 441 Instrumentation and ECT 453 Digital Computer for Process Control courses in the Electronics and Computer Technology Program (ECT) at the Bowling Green State University (BGSU). The paper first gives an overview of the current state of fieldbus networks in the industry. It lists various advantages of using fieldbus networks over point-to-point connections for instrumentation and control system implementations. The generic communication protocol model is discussed and the deviations from this model for various fieldbus networks are identified. As an example of a fieldbus, Controller Area Network (CAN) overview is presented. CANoe, a CAN simulation software is outlined. Details of a CAN bus based laboratory development at BGSU that uses CANoe software and the CAN hardware are also presented.

I. Introduction

Digital communication networks such as AS-I, CAN, Devicenet, Ethernet, Foundation Fieldbus, Profibus are increasingly used in instrumentation and control system implementations these days [1]. Sensors, controllers, and actuators are connected as nodes in these networks instead of hardwiring the devices with point-to-point connections. These networks, collectively called fieldbus networks, reduce system wiring and provide easy system diagnosis and maintenance. It is important to integrate fieldbus networks topic in instrumentation and control system courses in order to make the content of these courses up-to-date with the current industrial practice.

There is significant literature available on fieldbus networks [1-3]. Hulsebos has been maintaining a comprehensive web site since 1999 that lists various fieldbus networks with links to official web sites of each fieldbus organization [4]. Integration of fieldbus topics into undergraduate curriculum is slowly taking places at various institutions. For example, Franz [5] reported the development of a National Center for Digital and Fieldbus Technology (NCDFT) under an NSF grant at Lee College, Texas. Also in Reference [6], Müller and Max Felser described how fieldbus concepts are adopted in control technology curriculum in Switzerland. A weather station instrumentation experiment that uses digital and wireless communication concepts was adopted in a Computer Engineering curriculum at University of Oviedo, Spain [7]. The concept of fieldbus networks such as Devicenet are also introduced in PLC courses at several institutions [8-11]. Further more, some institutions such as University of Main revised its traditional power courses into industrial automation and communication courses [12]. An in depth understanding of the literature reveals that there is still a greater need to integrate fieldbus Page 11.642.2 topic into undergraduate engineering and technology curriculum.

This paper, therefore, presents how the fieldbus networks topic is integrated into ECT 441 Instrumentation and ECT 453 Digital Computer for Process Control courses in the ECT Program at BGSU. Section II describes the content of instrumentation and process control courses. Section III gives an overview of the current state of fieldbus networks in the industry. It lists various advantages of using fieldbus networks. The generic communication protocol model is discussed and the deviations from this model for various fieldbus networks are identified. An overview of CAN bus network is also given in that section. CANoe software, developed by Vector CANtech, is outlined in Section IV. Details of CAN bus based laboratory development at BGSU are presented in Section V. Concluding remarks are offered in Section VI.

II. Instrumentation and Process Control Courses at BGSU

The ECT program at BGSU offers instrumentation and process control courses in its undergraduate curriculum [13]. A graduate instrumentation and process control course is also offered in the Master of Industrial Technology program of the College of Technology. All of these courses have a laboratory component integrated with the lectures. The laboratory activities in these courses emphasize industrial sensors, actuators and data acquisition to investigate the behavior of the measurement and control systems. National Instrument’s NI-ELVIS station with LabVIEW software is used in these laboratory activities. Students do mini-projects using the PC-based laboratory workstations that integrate NI-ELVIS shown in Figure 1. These projects enable the students to analyze, design, build and test complete instrumentation and process control systems. Through this approach, students obtain exposure to many real problems associated with instruments such as loading, grounding, interaction between different blocks, nonlinearity, effects of ambient conditions and interfacing real world sensors and actuators to computers. Also, experience is gained in developing and testing process control algorithms using physical process simulators.

Page 11.642.3

Figure 1. NI-ELVIS Experimental Setup.

In the instrumentation course, some of the laboratory projects performed by students include i) temperature measurement system using RTD, ii) position measurement system using LVDT, and iii) weight measurement system using strain gauge. The projects performed in the process control course laboratory include, i) Temperature control using ON/OFF control mode, ii) PID controller for liquid level control system, and iii) speed control of dc motor. The temperature control system experiment described below is used as an example for the process control laboratory [14].

The objective of the temperature control experiment is to maintain the temperature inside a wooden box at some desired set-point value, within neutral zone limits, using a two-state- controller mode. The wooden box is heated with a light bulb. The temperature is measured using LM34 solid-state temperature sensor based circuit, which gives 10mV/°F. A signal conditioner is designed to modify the voltage corresponding to 50 to 150 °F to analog input voltage range of -10 V to +10 V. The output of the signal conditioner circuit is connected to an analog input channel on the NI-ELVIS station. When the temperature differs from the set-point value (with ± neutral zone), it results in the high-limit and the low-limit, and a fan is turned on and off accordingly. An analog output channel of the NI-ELVIS is connected to the solid-state relay (SSR) that controls the operation of the fan. Figure 2 gives a schematic diagram of this experiment.

Figure 2. Schematic Diagram of the Temperature Control Experiment.

A closer look at the content of the instrumentation and process control courses reveals a need to incorporate recent topics such a fieldbus networks. Also, current industrial employers are looking for employees with a background in fieldbus networks. It is therefore necessary to

provide some knowledge in fieldbus networks to our students. Page 11.642.4

III. Fieldbus Concepts

Fieldbus networks are digital communication networks in which sensors, controllers, and actuators are connected as nodes instead of hardwiring the devices with point-to-point connections. The advent of fieldbus technology has made possible a wide range of new capabilities. These new capabilities provide benefits that in turn translate into savings. The following are some of the advantages of fieldbus networks [1]:

1. Interoperability: the ability of a device to work together with other devices. This enables easy and tighter integration of devices from different manufacturers. 2. Greater system functionality: an unlimited number of parameters can be accessed from a device. This allows multiple measurements and a wide range of features to be crammed into advanced device firmware and accessed remotely by sophisticated software. 3. Simplicity: many features of fieldbus devices enhance their convenience and ease of use in comparison to traditional analog equipment. 4. Accuracy: Traditional 4-20 mA based systems require several stages of analog-to-digital (A/D) and digital-to-analog (D/A) conversions introducing quantization and other errors. Digital networking eliminates the need for these conversions. 5. Less cost of purchase: a system that is based on fieldbus technology requires significantly less hardware than a traditional system. 6. Savings: fieldbus based systems will have a) engineering savings, b) construction savings, c) maintenance savings, and d) operation savings. 7. Lower cost of expansion and change: Since fieldbus systems are cheaper to buy and deploy, they are also cheaper to expand and modify.

There are many fieldbus networks. Noel classified fieldbuses as discrete buses and process buses [15]. Discrete buses primarily focus in the discrete manufacturing area and are typically ON/OFF action, simple switches or low-level sensors. Competing discrete buses are: AS-I bus, Devicenet, Interbus-S, Profibus DP, SERCOS, Seriplex, and SDS. Process buses primarily focus in the process industry which predominantly uses modulating control. Competing process buses are FIP, Foundation Fieldbus, HART, LonWorks, Profibus-PA, and SP-50.

III.1 OSI/ISO Model

Today's network interconnection and interface models are framed in the context of the well- known Open Systems Interconnection (OSI) Model [16]. This model is the most influential network device communication model in existence. Conceived in the 1970s, the model was established in 1984 as the ISO standard 7498. The OSI model, shown in Figure 3, consists of seven interconnected layers and is arranged as a stack with the bottom layer physically connected to the network medium and the top layer accessible from its host's operating system. The top layer is known as the application layer. The presentation and session layers precede this layer. The transport and network layers precede those three high level layers. The bottom two layers are the data link and physical layers. Each layer is responsible for operating its services in a near simultaneous manner. Each layer must be able to communicate with its adjacent layers.

Page 11.642.5

Fieldbus networks do not implement all the seven layers. The network layer and the transport layer protocols are not implemented and these layers are omitted. The session, presentation and application layer are merged into a single application layer. With the missing middle layers, the second layer connects directly to the seventh layer. In Foundation Fieldbus, the application layer provides high-level functionality as defined by the Fieldbus Message Specification (FMS) and the Fieldbus Access Sublayer (FAS). A final layer known as the User Layer caps the Foundation Fieldbus HSE and H1 models and others. This additional eighth layer allows the fieldbus architecture to include necessary features such as a user-friendly interface.

Figure 3. ISO/OSI Seven Layer Model.

Currently the marketplace supports a large number of fieldbus vendors. Each of these vendors’ designs has their own strengths and weaknesses. Noel indicated that it is not easy to pick the right fieldbus [15]. The following points may be considered in selecting a suitable fieldbus:

1. Focus on the application far more than the technology. 2. Consider the costs. 3. Assess the network connectivity. 4. Understand the hidden changes and impacts. 5. Evaluate the real interoperability.

It is important to note that there is no single perfect fieldbus network, and the communication solutions for process control do not come in a “one size fits all” package. In the instrumentation and process control courses at BGSU, CAN bus topic is covered in detail as an example of a fieldbus. A brief account of CAN bus follows.

III.2 CAN Bus Overview

The Controller Area Network (CAN) standard, popular in automotive applications, defines a simple broadcast serial network that works well for real-time short-range communications [17- 19]. Bosch developed the CAN protocol, which has since been standardized internationally as ISO11898 and has been “implemented in silicon” by several semiconductor manufacturers. CAN is the basis of several sensor buses such as Devicenet, CANopen, J1939, and Smart Distributed System. Page 11.642.6

CAN uses a twisted pair cable to communicate up to 40m at speeds 1Mbit/s without repeaters, and up to 1 km at 20 kbps speed. It can support up to 40 devices. CAN uses CSMA bus arbitration. The CAN protocol, which corresponds to the data link and physical layers in the ISO/OSI reference model shown in Figure 3, meets the real-time requirements of automotive applications. CAN data packets are 8 bytes long and use 11-bit packet identifier. A second version of CAN can support 29 bit identifier.

Each CAN data frame consists of seven different bit fields shown in Figure 4. A data frame begins with the start-of-frame (SOF) bit. It is followed by an eleven-bit identifier and the remote transmission request (RTR) bit. The identifier and the RTR bit form the arbitration field. The control field consists of six bits and indicates how many bytes of data follow in the data field. The data field can be zero to eight bytes. The data field is followed by the cyclic redundancy checksum (CRC) field, which enables the receiver to check if the received bit sequence was corrupted. The two-bit acknowledgment (ACK) field is used by the transmitter to receive an acknowledgment of a valid frame from any receiver. The end of a message frame is signaled through a seven-bit end-of-frame (EOF). Further details of CAN such as arbitration and error handing can be found in many references [17].

Figure 4. CAN Data Frame.

IV. CANoe Simulation Software

CANoe, developed by Vector CANtech, is a robust CAN tool that is capable of simulating an entire CAN system [20]. CANoe supports the entire development process for networked systems from planning to implementation. CANoe offers special functions for all phases of the development process of distributed systems and its Electronic Control Units (ECUs), e.g. model creation, simulation, functional testing, diagnostics, and analysis.

CANoe supports the test of ECUs and networks via special functions of the Test Feature Set (TFS). With these functions, tests can be created to verify single development steps, check prototypes or execute regression and conformity tests. Additionally, the check and stimulus functions, included in the Test Service Library (TSL), simplify the setup and execution of its own test scenarios.

The CANoe functionality can be expanded as desired. Blocks may be inserted at any point in the data flow diagram, and its function can be programmed. The application-oriented, C-like language CAPL (Communication Access Programming Language) is used for this purpose. CANoe includes an interactive development environment, which makes it easy to create, modify, and compile CAPL programs. Page 11.642.7

Figure 5, reproduced from reference [20], shows the internal structure of CANoe. Network node models are added to the simulation setup as CAPL programs. These can be created manually or automatically from the database. The Panel Editor and Panel Generator support in creating graphic user control and display panels for the network node models. With system tests, it is often the case that peripheral signals of ECUs must be accessed. This is achieved by reading-in or outputting these signals over a port as environment variables, and these are used in the simulation.

A model for the network system can be created in CANoe using three steps:

1. Create the database with messages, signals and environment variables. CANdb++ editor is used to create these databases.

2. Create the network node periphery, which includes control panels. Panel Editor is used to create these panels.

3. Create the network node model in CAPL program.

Once the system is created, the program is executed and performance characteristics are monitored. The measurement setup windows such as trace window and statistics window allow the observation of these characteristics.

Figure 5. Internal Structure of CANoe (reproduced from reference [20]).

Page 11.642.8

V. Fieldbus Laboratory Exercises in Instrumentation and Control Courses Using CANoe

Several laboratory activities are being developed using the CANoe software. First an introductory lab exercise that acquaints students to CANoe environment is developed. In this lab exercise, students simulate a simple light bulb and switch. Students follow the three steps described in the previous section. They first create the database with signals and environmental variables using CANdb++. They then create the panels for the switch and light using the panel editor. Figure 6 shows these panels. Network node model, shown in Figure 7, is created using CAPL program. The simulation setup is then executed to see its operation. Whenever the switch is activated the indicator lamp illuminates. Whenever the switch is turned off, the indicator lamp goes off. The trace window shows both the bus communication, and the values of environment variables.

Figure 6. Measurement setup of CANoe.

A second laboratory activity simulates the temperature control experiment described in Section II above. The experiment that was performed before with point-to-point connections of temperature sensor, fan actuator, and PC controller with NI-ELVIS is simulated using CAN digital communication concepts. Figure 8 shows the CAN bus implementation of this system. It is being simulated using CANoe and its performance is being observed.

A third experiment simulates various control modules in an automobile using CANoe. These

include driver interface module, climate control module and door modules. This experiment is Page 11.642.9 under developmental stage at this time.

Figure 7. Simulation Setup of CANoe.

Figure 8. CAN Bus Implementation of Temperature Control System.

VI. Conclusions

Page 11.642.10 The purpose of this paper was to present the integration of fieldbus networks topic into instrumentation and process control courses offered by the ECT Program at BGSU. A review of

the content of these courses indicates a need to incorporate latest industrial developments. The current state of fieldbus networks in the industrial was first discussed. The generic ISO/OSI communication protocol model was described and the deviations from this model for various fieldbus networks were identified. As an example of a fieldbus network, an overview of CAN was given. CANoe software developed by Vector CANtech was outlined. Details of a CAN bus based laboratory development at BGSU were presented. These new developments will align the instrumentation and process control courses offered by the ECT program at BGSU with current industrial practices.

References

[1] Jonas Berge, Fieldbuses for Process Control: Engineering, Operation, and Maintenance , ISA - The Instrumentation, Systems and Automation Society Press, Research Triangle Park, NC, 2002. [2] G. Cena, L. Durante and A. Valenzano, “Evolution of standard fieldbus networks,” in Intelligent Systems and Robotics (G. W. Zobrist and C. Y. Ho, Eds.), Gordon and Beach Science Publisher, 2000. [3] Ian Akyildiz, Weilian Su, Yogesh Sankarasubramanian, and Erdal Cayirci, “A survey on Sensor Networks,” IEEE Communications Magazine, pp. 102-114, August 2002. [4] R. A. Hulsebos, “The fieldbus reference list,” http://ourworld.cs.com/_ht_a/rahulsebos/Links.htm , 1999-2005. [5] Harry Franz, “Fieldbus Instrumentation Technology Development at Houston, Texas Area Universities and Colleges,” Proceedings of ASEE Annual Conference, 2003. [6] Thomas Müller and Max Felser, “Adaptations to the embedded controls engineering curriculum in Switzerland,” 5th IFAC International Symposium on intelligent Components and Instruments for Control Applications (SICICA 2003), July 9-11, 2003, pp. 57-59

[7] Francisco Ferrero Martin, et al., “An Electronic Instrumentation Design Project for Computer Engineering Students,” IEEE Transactions on Education, vol. 48, pp. 472-481, August 2005. [8] W. Lin, et al., “Integration of enterprise and industrial networks in computer engineering technology program,” Proceedings of ASEE Annual Conference, 2004. [9] J. Rehg and B. Muller, “Teaching PLCs with the IEC 61131 standard languages,” Proceedings of ASEE Annual Conference, 2005. [10] J. Tapper, “Industry driven curriculum development, the key to successful courseware,” Proceedings of ASEE Annual Conference, 2001. [11] D. Wang, and H. Peddle, “System approach for design and construction of PLC training laboratory,” Proceedings of ASEE Annual Conference, 2001. [12] S. Dunning., “The revision of power courses into industrial automation and communication courses,” Proceedings of ASEE Annual Conference, 2002. [13] Curtis Johnson, Process Control Instrumentation Technology , Prentice Hall, 2006. [14] Edwin Rézaei and Sri Kolla., “Internet-based ON/OFF controller using LabVIEW,” Proceedings of ASEE Annual Conference, 2003. [15] J. Noel, “Digital bus selection – It’s not your typical quick pick at 7-11,” Proceedings of 2002 ISA EXPO, Chicago, IL, October 2002. [16] Sri Kolla, et al., “Fieldbus networks for control system implementations,” Proceedings of EIC/EMCWA EXPO, 2003. [17] K. H. Johansson, et al. (2005). “Vehicle applications of Controller Area Networks.” To appear in 2005 in the Handbook of Networked and Embedded Control Systems, D. Hristu-Varsakelis and W. S. Levine, Eds., Retrieved January 19, 2006 from http://www.md.kth.se/RTC/Papers/VehicleApplicationsCan2005.pdf . [18] H. Kopetz, “A comparison of CAN and TTP.” Retrieved on January 19, 2006 from http://www.md.kth.se/RTC/RTCC/Material01/Papers/kopetz-ttpvscan.pdf. [19] CAN protocol training notes, Vector CANtech Inc., 2004. [20] CANoe Manual, Version 5.1.1, Vector CANtech Inc., 2005. Page 11.642.11